{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1949,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 1.7453284027716789,
      "learning_rate": 1.0256410256410257e-06,
      "loss": 4.8029,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.8838630132500438,
      "learning_rate": 5.128205128205128e-06,
      "loss": 4.8678,
      "step": 5
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.9750863683254836,
      "learning_rate": 1.0256410256410256e-05,
      "loss": 4.7915,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.0960739711216374,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 4.7429,
      "step": 15
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.68908841814168,
      "learning_rate": 2.0512820512820512e-05,
      "loss": 4.655,
      "step": 20
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.4517244544448675,
      "learning_rate": 2.564102564102564e-05,
      "loss": 4.3514,
      "step": 25
    },
    {
      "epoch": 0.02,
      "grad_norm": 4.136057917635942,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 3.905,
      "step": 30
    },
    {
      "epoch": 0.02,
      "grad_norm": 4.589117875619624,
      "learning_rate": 3.58974358974359e-05,
      "loss": 3.2509,
      "step": 35
    },
    {
      "epoch": 0.02,
      "grad_norm": 4.927853133671092,
      "learning_rate": 4.1025641025641023e-05,
      "loss": 2.6795,
      "step": 40
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.054252767001273,
      "learning_rate": 4.615384615384616e-05,
      "loss": 2.187,
      "step": 45
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.9849933991408988,
      "learning_rate": 5.128205128205128e-05,
      "loss": 1.9199,
      "step": 50
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.8072063144215404,
      "learning_rate": 5.6410256410256414e-05,
      "loss": 1.7439,
      "step": 55
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.3923291097565402,
      "learning_rate": 6.153846153846155e-05,
      "loss": 1.7219,
      "step": 60
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.1768172611689185,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1.6276,
      "step": 65
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.1476923157725187,
      "learning_rate": 7.17948717948718e-05,
      "loss": 1.5326,
      "step": 70
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.9441328628031238,
      "learning_rate": 7.692307692307693e-05,
      "loss": 1.5766,
      "step": 75
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8493903573919569,
      "learning_rate": 8.205128205128205e-05,
      "loss": 1.4017,
      "step": 80
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.0349555509222987,
      "learning_rate": 8.717948717948718e-05,
      "loss": 1.447,
      "step": 85
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.8454953032919131,
      "learning_rate": 9.230769230769232e-05,
      "loss": 1.4113,
      "step": 90
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.0220924006013166,
      "learning_rate": 9.743589743589744e-05,
      "loss": 1.4865,
      "step": 95
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.9739811152642958,
      "learning_rate": 0.00010256410256410256,
      "loss": 1.4709,
      "step": 100
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.8989088188815814,
      "learning_rate": 0.0001076923076923077,
      "loss": 1.4335,
      "step": 105
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.9043247030909609,
      "learning_rate": 0.00011282051282051283,
      "loss": 1.4196,
      "step": 110
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.0554112608560857,
      "learning_rate": 0.00011794871794871796,
      "loss": 1.376,
      "step": 115
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.9282876975712975,
      "learning_rate": 0.0001230769230769231,
      "loss": 1.4137,
      "step": 120
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.8957189154484865,
      "learning_rate": 0.00012820512820512823,
      "loss": 1.3528,
      "step": 125
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.3941737136229102,
      "learning_rate": 0.00013333333333333334,
      "loss": 1.3963,
      "step": 130
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.7436200394991389,
      "learning_rate": 0.00013846153846153847,
      "loss": 1.3047,
      "step": 135
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.1477489067913107,
      "learning_rate": 0.0001435897435897436,
      "loss": 1.3011,
      "step": 140
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.8722003799830739,
      "learning_rate": 0.00014871794871794872,
      "loss": 1.2378,
      "step": 145
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.1769840880132467,
      "learning_rate": 0.00015384615384615385,
      "loss": 1.4124,
      "step": 150
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.7399233026092019,
      "learning_rate": 0.00015897435897435896,
      "loss": 1.3027,
      "step": 155
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.9538756518623174,
      "learning_rate": 0.0001641025641025641,
      "loss": 1.2812,
      "step": 160
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.987821734343656,
      "learning_rate": 0.00016923076923076923,
      "loss": 1.3357,
      "step": 165
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.9972992684207476,
      "learning_rate": 0.00017435897435897436,
      "loss": 1.3845,
      "step": 170
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.9162912417797362,
      "learning_rate": 0.0001794871794871795,
      "loss": 1.4145,
      "step": 175
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.9727023032566935,
      "learning_rate": 0.00018461538461538463,
      "loss": 1.2514,
      "step": 180
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.1205678233237835,
      "learning_rate": 0.00018974358974358974,
      "loss": 1.3525,
      "step": 185
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.9769806404545222,
      "learning_rate": 0.00019487179487179487,
      "loss": 1.2291,
      "step": 190
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.8933569683133975,
      "learning_rate": 0.0002,
      "loss": 1.3543,
      "step": 195
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.9689935176055495,
      "learning_rate": 0.00019999598996948235,
      "loss": 1.2744,
      "step": 200
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.9199321088853798,
      "learning_rate": 0.00019998396019953624,
      "loss": 1.3059,
      "step": 205
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.0113198660825118,
      "learning_rate": 0.0001999639116549566,
      "loss": 1.1999,
      "step": 210
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.7980098887264342,
      "learning_rate": 0.00019993584594364894,
      "loss": 1.2418,
      "step": 215
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.7449325030003408,
      "learning_rate": 0.0001998997653165004,
      "loss": 1.284,
      "step": 220
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.0156080824941,
      "learning_rate": 0.00019985567266719934,
      "loss": 1.3727,
      "step": 225
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.0387246660942182,
      "learning_rate": 0.00019980357153200315,
      "loss": 1.3186,
      "step": 230
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.1886194935509105,
      "learning_rate": 0.00019974346608945466,
      "loss": 1.2949,
      "step": 235
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.9654192561255139,
      "learning_rate": 0.00019967536116004698,
      "loss": 1.3067,
      "step": 240
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.0385716215564977,
      "learning_rate": 0.00019959926220583713,
      "loss": 1.289,
      "step": 245
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.8709732110452237,
      "learning_rate": 0.00019951517533000764,
      "loss": 1.2138,
      "step": 250
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.6847073355471284,
      "learning_rate": 0.00019942310727637724,
      "loss": 1.2844,
      "step": 255
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.051169320358427,
      "learning_rate": 0.00019932306542886009,
      "loss": 1.3289,
      "step": 260
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.8704314065526715,
      "learning_rate": 0.00019921505781087334,
      "loss": 1.3339,
      "step": 265
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.7816111152013846,
      "learning_rate": 0.00019909909308469398,
      "loss": 1.3282,
      "step": 270
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.038569084884339,
      "learning_rate": 0.0001989751805507637,
      "loss": 1.3112,
      "step": 275
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.9061282430692424,
      "learning_rate": 0.00019884333014694345,
      "loss": 1.3251,
      "step": 280
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.7513521845340044,
      "learning_rate": 0.00019870355244771607,
      "loss": 1.2893,
      "step": 285
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.910789847545989,
      "learning_rate": 0.00019855585866333835,
      "loss": 1.288,
      "step": 290
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.0532355924793768,
      "learning_rate": 0.00019840026063894193,
      "loss": 1.2805,
      "step": 295
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.0244686274469181,
      "learning_rate": 0.00019823677085358335,
      "loss": 1.2499,
      "step": 300
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7707389186864114,
      "learning_rate": 0.00019806540241924317,
      "loss": 1.2532,
      "step": 305
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7446042971478654,
      "learning_rate": 0.00019788616907977441,
      "loss": 1.2981,
      "step": 310
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6848536091053609,
      "learning_rate": 0.00019769908520980034,
      "loss": 1.2565,
      "step": 315
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7586642653181629,
      "learning_rate": 0.00019750416581356146,
      "loss": 1.2648,
      "step": 320
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.9291629579523186,
      "learning_rate": 0.00019730142652371236,
      "loss": 1.1065,
      "step": 325
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.7774039891171479,
      "learning_rate": 0.0001970908836000678,
      "loss": 1.2582,
      "step": 330
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.7535737744178963,
      "learning_rate": 0.00019687255392829877,
      "loss": 1.255,
      "step": 335
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.8232038564451061,
      "learning_rate": 0.0001966464550185782,
      "loss": 1.2246,
      "step": 340
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.7763354984102764,
      "learning_rate": 0.0001964126050041767,
      "loss": 1.1932,
      "step": 345
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.9765706941888299,
      "learning_rate": 0.0001961710226400081,
      "loss": 1.2209,
      "step": 350
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.7182663624300912,
      "learning_rate": 0.00019592172730112544,
      "loss": 1.2383,
      "step": 355
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.8481687196291695,
      "learning_rate": 0.00019566473898116713,
      "loss": 1.2911,
      "step": 360
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.7711940010123207,
      "learning_rate": 0.0001954000782907532,
      "loss": 1.231,
      "step": 365
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.061067518571626,
      "learning_rate": 0.00019512776645583263,
      "loss": 1.259,
      "step": 370
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.0715949890319147,
      "learning_rate": 0.00019484782531598073,
      "loss": 1.2712,
      "step": 375
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.774154512584797,
      "learning_rate": 0.00019456027732264784,
      "loss": 1.306,
      "step": 380
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.8172535402599491,
      "learning_rate": 0.00019426514553735848,
      "loss": 1.2734,
      "step": 385
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.9451921871881396,
      "learning_rate": 0.00019396245362986197,
      "loss": 1.231,
      "step": 390
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.9954188288045217,
      "learning_rate": 0.00019365222587623405,
      "loss": 1.2061,
      "step": 395
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.8088092576041586,
      "learning_rate": 0.00019333448715692995,
      "loss": 1.2403,
      "step": 400
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.9867132889317975,
      "learning_rate": 0.00019300926295478884,
      "loss": 1.3009,
      "step": 405
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.8580342036883377,
      "learning_rate": 0.0001926765793529902,
      "loss": 1.2371,
      "step": 410
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.813693213024289,
      "learning_rate": 0.00019233646303296205,
      "loss": 1.2767,
      "step": 415
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.8552421430267484,
      "learning_rate": 0.00019198894127224074,
      "loss": 1.2419,
      "step": 420
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.9962550179351269,
      "learning_rate": 0.0001916340419422837,
      "loss": 1.2765,
      "step": 425
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.790052217945725,
      "learning_rate": 0.00019127179350623372,
      "loss": 1.2135,
      "step": 430
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.7916779759677229,
      "learning_rate": 0.0001909022250166365,
      "loss": 1.2178,
      "step": 435
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.7297942006635301,
      "learning_rate": 0.00019052536611311046,
      "loss": 1.245,
      "step": 440
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.8857053212429089,
      "learning_rate": 0.00019014124701996973,
      "loss": 1.2657,
      "step": 445
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.726328154530024,
      "learning_rate": 0.00018974989854379996,
      "loss": 1.2481,
      "step": 450
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.0011198664456165,
      "learning_rate": 0.00018935135207098785,
      "loss": 1.2323,
      "step": 455
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.7783715496305177,
      "learning_rate": 0.00018894563956520374,
      "loss": 1.2367,
      "step": 460
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.8696181443757303,
      "learning_rate": 0.00018853279356483826,
      "loss": 1.315,
      "step": 465
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.7894223278767809,
      "learning_rate": 0.00018811284718039256,
      "loss": 1.124,
      "step": 470
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.1136345673311567,
      "learning_rate": 0.00018768583409182305,
      "loss": 1.2198,
      "step": 475
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.7243107301552627,
      "learning_rate": 0.00018725178854584007,
      "loss": 1.2523,
      "step": 480
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.6223206270124363,
      "learning_rate": 0.00018681074535316125,
      "loss": 1.2673,
      "step": 485
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.8039914099462059,
      "learning_rate": 0.00018636273988571991,
      "loss": 1.1877,
      "step": 490
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.7116186179251531,
      "learning_rate": 0.0001859078080738279,
      "loss": 1.1458,
      "step": 495
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.7310411784585837,
      "learning_rate": 0.00018544598640329432,
      "loss": 1.2417,
      "step": 500
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.7429734188864324,
      "learning_rate": 0.00018497731191249894,
      "loss": 1.2852,
      "step": 505
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.9083872621008465,
      "learning_rate": 0.000184501822189422,
      "loss": 1.2436,
      "step": 510
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.9414190517105039,
      "learning_rate": 0.00018401955536862948,
      "loss": 1.2382,
      "step": 515
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.9100118313504707,
      "learning_rate": 0.0001835305501282148,
      "loss": 1.2651,
      "step": 520
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.8682421638749216,
      "learning_rate": 0.00018303484568669667,
      "loss": 1.214,
      "step": 525
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.7124640006090736,
      "learning_rate": 0.00018253248179987388,
      "loss": 1.1703,
      "step": 530
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.7882268918908124,
      "learning_rate": 0.0001820234987576368,
      "loss": 1.1747,
      "step": 535
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.9355156994606193,
      "learning_rate": 0.00018150793738073602,
      "loss": 1.1954,
      "step": 540
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.962953079362863,
      "learning_rate": 0.00018098583901750867,
      "loss": 1.21,
      "step": 545
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.8058576401677449,
      "learning_rate": 0.00018045724554056214,
      "loss": 1.1945,
      "step": 550
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.8718736747079745,
      "learning_rate": 0.0001799221993434159,
      "loss": 1.242,
      "step": 555
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.8820093361979019,
      "learning_rate": 0.00017938074333710157,
      "loss": 1.216,
      "step": 560
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.626929459625588,
      "learning_rate": 0.00017883292094672128,
      "loss": 1.1947,
      "step": 565
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.8252467152886114,
      "learning_rate": 0.00017827877610796514,
      "loss": 1.1953,
      "step": 570
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.894675298458733,
      "learning_rate": 0.00017771835326358743,
      "loss": 1.1873,
      "step": 575
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6648245020030112,
      "learning_rate": 0.00017715169735984233,
      "loss": 1.1293,
      "step": 580
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.8629378772284809,
      "learning_rate": 0.0001765788538428792,
      "loss": 1.288,
      "step": 585
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.9066125782712056,
      "learning_rate": 0.00017599986865509767,
      "loss": 1.2058,
      "step": 590
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.7423874071094034,
      "learning_rate": 0.00017541478823146327,
      "loss": 1.2508,
      "step": 595
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.01835592219078,
      "learning_rate": 0.00017482365949578302,
      "loss": 1.1974,
      "step": 600
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.878167603852895,
      "learning_rate": 0.00017422652985694237,
      "loss": 1.2583,
      "step": 605
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.7708769505928975,
      "learning_rate": 0.00017362344720510278,
      "loss": 1.1853,
      "step": 610
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.7652211411520762,
      "learning_rate": 0.00017301445990786102,
      "loss": 1.2707,
      "step": 615
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.7286424011748441,
      "learning_rate": 0.00017239961680637,
      "loss": 1.2044,
      "step": 620
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.6950598836326154,
      "learning_rate": 0.0001717789672114218,
      "loss": 1.3171,
      "step": 625
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.7009337184084006,
      "learning_rate": 0.0001711525608994927,
      "loss": 1.1541,
      "step": 630
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.8543565184004501,
      "learning_rate": 0.00017052044810875126,
      "loss": 1.2465,
      "step": 635
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.1754223407681135,
      "learning_rate": 0.00016988267953502913,
      "loss": 1.1644,
      "step": 640
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.8933582330947162,
      "learning_rate": 0.00016923930632775516,
      "loss": 1.1798,
      "step": 645
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.932416979264158,
      "learning_rate": 0.00016859038008585326,
      "loss": 1.1983,
      "step": 650
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.8373224826856496,
      "learning_rate": 0.0001679359528536041,
      "loss": 1.1567,
      "step": 655
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.8214365574255561,
      "learning_rate": 0.00016727607711647114,
      "loss": 1.2188,
      "step": 660
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.9351111268728677,
      "learning_rate": 0.00016661080579689132,
      "loss": 1.1873,
      "step": 665
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.8498925724649622,
      "learning_rate": 0.0001659401922500304,
      "loss": 1.1937,
      "step": 670
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.7284767681268423,
      "learning_rate": 0.00016526429025950424,
      "loss": 1.1301,
      "step": 675
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.7481373744501812,
      "learning_rate": 0.00016458315403306502,
      "loss": 1.2671,
      "step": 680
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.8920310346592031,
      "learning_rate": 0.0001638968381982538,
      "loss": 1.1528,
      "step": 685
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.7925520466506051,
      "learning_rate": 0.0001632053977980194,
      "loss": 1.2368,
      "step": 690
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.0192425967562995,
      "learning_rate": 0.000162508888286304,
      "loss": 1.2167,
      "step": 695
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.7085398144677333,
      "learning_rate": 0.00016180736552359553,
      "loss": 1.2218,
      "step": 700
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.768972223449797,
      "learning_rate": 0.00016110088577244773,
      "loss": 1.2483,
      "step": 705
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.8053989910048416,
      "learning_rate": 0.00016038950569296785,
      "loss": 1.1811,
      "step": 710
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.7498380291821676,
      "learning_rate": 0.00015967328233827249,
      "loss": 1.2625,
      "step": 715
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.7372396772067858,
      "learning_rate": 0.00015895227314991178,
      "loss": 1.1298,
      "step": 720
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.7380540614428072,
      "learning_rate": 0.00015822653595326275,
      "loss": 1.1611,
      "step": 725
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.840298145620476,
      "learning_rate": 0.00015749612895289152,
      "loss": 1.1815,
      "step": 730
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.6743184976694755,
      "learning_rate": 0.00015676111072788527,
      "loss": 1.1665,
      "step": 735
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.0773509553134593,
      "learning_rate": 0.00015602154022715435,
      "loss": 1.214,
      "step": 740
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.2157724994677412,
      "learning_rate": 0.0001552774767647043,
      "loss": 1.1757,
      "step": 745
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.8932509973388469,
      "learning_rate": 0.0001545289800148789,
      "loss": 1.1747,
      "step": 750
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.773624653232096,
      "learning_rate": 0.0001537761100075744,
      "loss": 1.1934,
      "step": 755
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.6455277482504939,
      "learning_rate": 0.00015301892712342482,
      "loss": 1.2505,
      "step": 760
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.7996020026859186,
      "learning_rate": 0.00015225749208895968,
      "loss": 1.2388,
      "step": 765
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.7876699542363711,
      "learning_rate": 0.0001514918659717335,
      "loss": 1.2573,
      "step": 770
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.7507865451214197,
      "learning_rate": 0.00015072211017542813,
      "loss": 1.2108,
      "step": 775
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.742160262755694,
      "learning_rate": 0.00014994828643492827,
      "loss": 1.1554,
      "step": 780
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.7898189151931231,
      "learning_rate": 0.00014917045681137026,
      "loss": 1.1858,
      "step": 785
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.7929109677821212,
      "learning_rate": 0.0001483886836871646,
      "loss": 1.1181,
      "step": 790
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.8366860762380138,
      "learning_rate": 0.00014760302976099304,
      "loss": 1.1683,
      "step": 795
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.8562461991337411,
      "learning_rate": 0.00014681355804278001,
      "loss": 1.1545,
      "step": 800
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.6903144224243607,
      "learning_rate": 0.00014602033184863913,
      "loss": 1.1668,
      "step": 805
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.720558400777453,
      "learning_rate": 0.00014522341479579533,
      "loss": 1.3024,
      "step": 810
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.6374665332941237,
      "learning_rate": 0.00014442287079748263,
      "loss": 1.1983,
      "step": 815
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.8700325415003359,
      "learning_rate": 0.00014361876405781832,
      "loss": 1.1604,
      "step": 820
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.876670415809109,
      "learning_rate": 0.00014281115906665374,
      "loss": 1.2463,
      "step": 825
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.9136860273140437,
      "learning_rate": 0.00014200012059440207,
      "loss": 1.1814,
      "step": 830
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.8346225178534373,
      "learning_rate": 0.00014118571368684383,
      "loss": 1.2552,
      "step": 835
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.698383520911491,
      "learning_rate": 0.00014036800365991008,
      "loss": 1.1132,
      "step": 840
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.8620628308046092,
      "learning_rate": 0.00013954705609444404,
      "loss": 1.1663,
      "step": 845
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.8119834436603902,
      "learning_rate": 0.00013872293683094152,
      "loss": 1.2139,
      "step": 850
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.7204007887604488,
      "learning_rate": 0.00013789571196427055,
      "loss": 1.193,
      "step": 855
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.9461103539326526,
      "learning_rate": 0.00013706544783837022,
      "loss": 1.1549,
      "step": 860
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.1236203572833683,
      "learning_rate": 0.00013623221104093025,
      "loss": 1.2752,
      "step": 865
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.7297339296275664,
      "learning_rate": 0.00013539606839805036,
      "loss": 1.1884,
      "step": 870
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.7684475171790612,
      "learning_rate": 0.00013455708696888085,
      "loss": 1.1928,
      "step": 875
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.7160993458180628,
      "learning_rate": 0.00013371533404024438,
      "loss": 1.2035,
      "step": 880
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.7366548665049109,
      "learning_rate": 0.00013287087712123962,
      "loss": 1.2372,
      "step": 885
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.7983969367173553,
      "learning_rate": 0.00013202378393782692,
      "loss": 1.1931,
      "step": 890
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.9470935523528393,
      "learning_rate": 0.00013117412242739655,
      "loss": 1.1274,
      "step": 895
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.0451209845718006,
      "learning_rate": 0.00013032196073332027,
      "loss": 1.2435,
      "step": 900
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.7471749313716822,
      "learning_rate": 0.00012946736719948607,
      "loss": 1.1844,
      "step": 905
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.6962269357662448,
      "learning_rate": 0.000128610410364817,
      "loss": 1.1526,
      "step": 910
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.8391163485957552,
      "learning_rate": 0.00012775115895777417,
      "loss": 1.2126,
      "step": 915
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.9175047450119641,
      "learning_rate": 0.00012688968189084493,
      "loss": 1.1199,
      "step": 920
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.8881533984263993,
      "learning_rate": 0.00012602604825501587,
      "loss": 1.1842,
      "step": 925
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.68829893427438,
      "learning_rate": 0.00012516032731423165,
      "loss": 1.1916,
      "step": 930
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.8290955836420092,
      "learning_rate": 0.00012429258849984014,
      "loss": 1.1915,
      "step": 935
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.7546478279579057,
      "learning_rate": 0.00012342290140502388,
      "loss": 1.1875,
      "step": 940
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.7857053878657014,
      "learning_rate": 0.00012255133577921868,
      "loss": 1.1883,
      "step": 945
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.7728388731432995,
      "learning_rate": 0.0001216779615225197,
      "loss": 1.1329,
      "step": 950
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.762285378503341,
      "learning_rate": 0.00012080284868007541,
      "loss": 1.1857,
      "step": 955
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.8455730497463988,
      "learning_rate": 0.0001199260674364699,
      "loss": 1.2241,
      "step": 960
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.7637030076266145,
      "learning_rate": 0.00011904768811009405,
      "loss": 1.2333,
      "step": 965
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.7399802799801922,
      "learning_rate": 0.00011816778114750593,
      "loss": 1.2126,
      "step": 970
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.7804083472026294,
      "learning_rate": 0.00011728641711778103,
      "loss": 1.2239,
      "step": 975
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.7666274480697189,
      "learning_rate": 0.00011640366670685248,
      "loss": 1.105,
      "step": 980
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.7651209631784908,
      "learning_rate": 0.00011551960071184195,
      "loss": 1.3045,
      "step": 985
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.9726990471095971,
      "learning_rate": 0.00011463429003538196,
      "loss": 1.1956,
      "step": 990
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.7087953125740096,
      "learning_rate": 0.000113747805679929,
      "loss": 1.2197,
      "step": 995
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.8944588834117093,
      "learning_rate": 0.00011286021874206952,
      "loss": 1.0926,
      "step": 1000
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.0048773208945307,
      "learning_rate": 0.00011197160040681762,
      "loss": 1.2798,
      "step": 1005
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.7348160099574793,
      "learning_rate": 0.0001110820219419062,
      "loss": 1.1951,
      "step": 1010
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.7992861590760896,
      "learning_rate": 0.0001101915546920711,
      "loss": 1.1812,
      "step": 1015
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.7458148092571341,
      "learning_rate": 0.00010930027007332923,
      "loss": 1.1236,
      "step": 1020
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.7657904989795685,
      "learning_rate": 0.00010840823956725103,
      "loss": 1.1714,
      "step": 1025
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.7797801904462873,
      "learning_rate": 0.00010751553471522757,
      "loss": 1.164,
      "step": 1030
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.6805580259070753,
      "learning_rate": 0.00010662222711273279,
      "loss": 1.1326,
      "step": 1035
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.8265666935073783,
      "learning_rate": 0.00010572838840358168,
      "loss": 1.1233,
      "step": 1040
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.824127124532494,
      "learning_rate": 0.00010483409027418425,
      "loss": 1.1273,
      "step": 1045
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.8666956799033239,
      "learning_rate": 0.00010393940444779635,
      "loss": 1.1172,
      "step": 1050
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.7597618280586041,
      "learning_rate": 0.00010304440267876727,
      "loss": 1.2262,
      "step": 1055
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.8594584043580179,
      "learning_rate": 0.00010214915674678523,
      "loss": 1.237,
      "step": 1060
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.8503111582046956,
      "learning_rate": 0.00010125373845112034,
      "loss": 1.2224,
      "step": 1065
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.8459974040134323,
      "learning_rate": 0.00010035821960486643,
      "loss": 1.1968,
      "step": 1070
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.8236420083982484,
      "learning_rate": 9.946267202918157e-05,
      "loss": 1.1334,
      "step": 1075
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.9248108629370668,
      "learning_rate": 9.856716754752796e-05,
      "loss": 1.1352,
      "step": 1080
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.8808986018254903,
      "learning_rate": 9.767177797991155e-05,
      "loss": 1.1959,
      "step": 1085
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.635409279473428,
      "learning_rate": 9.677657513712221e-05,
      "loss": 1.1506,
      "step": 1090
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.7550156809913564,
      "learning_rate": 9.588163081497427e-05,
      "loss": 1.2308,
      "step": 1095
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.7609369024863887,
      "learning_rate": 9.498701678854865e-05,
      "loss": 1.1722,
      "step": 1100
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.8064136692370791,
      "learning_rate": 9.409280480643628e-05,
      "loss": 1.2279,
      "step": 1105
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.7131735035318008,
      "learning_rate": 9.319906658498389e-05,
      "loss": 1.1862,
      "step": 1110
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.8678942495261857,
      "learning_rate": 9.230587380254237e-05,
      "loss": 1.1531,
      "step": 1115
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.6604664522870226,
      "learning_rate": 9.141329809371803e-05,
      "loss": 1.1744,
      "step": 1120
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.7095675124068804,
      "learning_rate": 9.052141104362748e-05,
      "loss": 1.1311,
      "step": 1125
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.7845862314567627,
      "learning_rate": 8.963028418215653e-05,
      "loss": 1.2801,
      "step": 1130
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.8019276427700621,
      "learning_rate": 8.873998897822336e-05,
      "loss": 1.2425,
      "step": 1135
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.0094066794890815,
      "learning_rate": 8.785059683404672e-05,
      "loss": 1.1287,
      "step": 1140
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.9258556140522551,
      "learning_rate": 8.696217907941941e-05,
      "loss": 1.2059,
      "step": 1145
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.9454158193155473,
      "learning_rate": 8.607480696598762e-05,
      "loss": 1.0897,
      "step": 1150
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.8742232040464293,
      "learning_rate": 8.518855166153644e-05,
      "loss": 1.2234,
      "step": 1155
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.8764506057333485,
      "learning_rate": 8.43034842442822e-05,
      "loss": 1.1409,
      "step": 1160
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.9294531011673496,
      "learning_rate": 8.341967569717202e-05,
      "loss": 1.0986,
      "step": 1165
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.8413972427762672,
      "learning_rate": 8.253719690219079e-05,
      "loss": 1.1334,
      "step": 1170
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.9118769910143452,
      "learning_rate": 8.165611863467644e-05,
      "loss": 1.145,
      "step": 1175
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.7880447725035846,
      "learning_rate": 8.077651155764387e-05,
      "loss": 1.1299,
      "step": 1180
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.8382287781274999,
      "learning_rate": 7.98984462161175e-05,
      "loss": 1.2243,
      "step": 1185
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.6697229577393602,
      "learning_rate": 7.902199303147363e-05,
      "loss": 1.2093,
      "step": 1190
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.7808818157955248,
      "learning_rate": 7.814722229579264e-05,
      "loss": 1.1984,
      "step": 1195
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.9135627361643314,
      "learning_rate": 7.727420416622144e-05,
      "loss": 1.2091,
      "step": 1200
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.680528612137634,
      "learning_rate": 7.640300865934687e-05,
      "loss": 1.1034,
      "step": 1205
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.812190629178745,
      "learning_rate": 7.553370564558032e-05,
      "loss": 1.164,
      "step": 1210
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.6786859157828664,
      "learning_rate": 7.46663648435541e-05,
      "loss": 1.1711,
      "step": 1215
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.0608937731498502,
      "learning_rate": 7.380105581452987e-05,
      "loss": 1.1321,
      "step": 1220
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.8774620996995025,
      "learning_rate": 7.293784795681994e-05,
      "loss": 1.2073,
      "step": 1225
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.8397584969858537,
      "learning_rate": 7.207681050022132e-05,
      "loss": 1.1356,
      "step": 1230
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.7565443884452981,
      "learning_rate": 7.121801250046363e-05,
      "loss": 1.2799,
      "step": 1235
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.6473890021488468,
      "learning_rate": 7.036152283367056e-05,
      "loss": 1.1355,
      "step": 1240
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.8635483831775488,
      "learning_rate": 6.950741019083617e-05,
      "loss": 1.172,
      "step": 1245
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.7635797502403529,
      "learning_rate": 6.865574307231575e-05,
      "loss": 1.1359,
      "step": 1250
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.9134433746313595,
      "learning_rate": 6.780658978233199e-05,
      "loss": 1.1523,
      "step": 1255
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.7862648398970986,
      "learning_rate": 6.696001842349702e-05,
      "loss": 1.1786,
      "step": 1260
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.8737779184120746,
      "learning_rate": 6.611609689135056e-05,
      "loss": 1.2463,
      "step": 1265
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.8231816299971835,
      "learning_rate": 6.527489286891459e-05,
      "loss": 1.0201,
      "step": 1270
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.833558803199825,
      "learning_rate": 6.443647382126509e-05,
      "loss": 1.1555,
      "step": 1275
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.7866154882582159,
      "learning_rate": 6.360090699012145e-05,
      "loss": 1.138,
      "step": 1280
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.8309461290217522,
      "learning_rate": 6.27682593884535e-05,
      "loss": 1.0762,
      "step": 1285
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.7270575882509663,
      "learning_rate": 6.193859779510712e-05,
      "loss": 1.2055,
      "step": 1290
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.7267387499731184,
      "learning_rate": 6.111198874944845e-05,
      "loss": 1.1906,
      "step": 1295
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.8081247294023348,
      "learning_rate": 6.0288498546027536e-05,
      "loss": 1.2243,
      "step": 1300
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.8237165040439475,
      "learning_rate": 5.946819322926127e-05,
      "loss": 1.0994,
      "step": 1305
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.860921836417512,
      "learning_rate": 5.865113858813673e-05,
      "loss": 1.1615,
      "step": 1310
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.7995623045168802,
      "learning_rate": 5.783740015093484e-05,
      "loss": 1.1728,
      "step": 1315
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.7326482406819708,
      "learning_rate": 5.702704317997492e-05,
      "loss": 1.112,
      "step": 1320
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.8123434824427339,
      "learning_rate": 5.6220132666380635e-05,
      "loss": 1.0999,
      "step": 1325
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.8440723192599834,
      "learning_rate": 5.541673332486773e-05,
      "loss": 1.2143,
      "step": 1330
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.7328759482460477,
      "learning_rate": 5.4616909588553674e-05,
      "loss": 1.2724,
      "step": 1335
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.6946255004506102,
      "learning_rate": 5.3820725603790346e-05,
      "loss": 1.2365,
      "step": 1340
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.7864322469951978,
      "learning_rate": 5.30282452250193e-05,
      "loss": 1.2184,
      "step": 1345
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.7860065077827799,
      "learning_rate": 5.223953200965055e-05,
      "loss": 1.177,
      "step": 1350
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.7870610754641466,
      "learning_rate": 5.145464921296537e-05,
      "loss": 1.2478,
      "step": 1355
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.8098199890971101,
      "learning_rate": 5.067365978304315e-05,
      "loss": 1.1132,
      "step": 1360
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.7277674932711228,
      "learning_rate": 4.9896626355712805e-05,
      "loss": 1.1065,
      "step": 1365
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.6702441184475941,
      "learning_rate": 4.912361124952948e-05,
      "loss": 1.1489,
      "step": 1370
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.7382556968521137,
      "learning_rate": 4.835467646077656e-05,
      "loss": 1.1145,
      "step": 1375
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.7012581002313999,
      "learning_rate": 4.7589883658493296e-05,
      "loss": 1.1196,
      "step": 1380
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.8448006051293048,
      "learning_rate": 4.682929417952939e-05,
      "loss": 1.1081,
      "step": 1385
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.6907481068653331,
      "learning_rate": 4.6072969023625165e-05,
      "loss": 1.1438,
      "step": 1390
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.7998262194855656,
      "learning_rate": 4.532096884851978e-05,
      "loss": 1.1514,
      "step": 1395
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.9014262005578314,
      "learning_rate": 4.457335396508631e-05,
      "loss": 1.1583,
      "step": 1400
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.8456699239379848,
      "learning_rate": 4.383018433249464e-05,
      "loss": 1.1471,
      "step": 1405
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.7936192686618659,
      "learning_rate": 4.309151955340297e-05,
      "loss": 1.1212,
      "step": 1410
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.8376082121769935,
      "learning_rate": 4.2357418869177354e-05,
      "loss": 1.2077,
      "step": 1415
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.8095128041860435,
      "learning_rate": 4.162794115514078e-05,
      "loss": 1.2169,
      "step": 1420
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.7509614581437917,
      "learning_rate": 4.0903144915851174e-05,
      "loss": 1.0553,
      "step": 1425
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.7839995425652196,
      "learning_rate": 4.018308828040924e-05,
      "loss": 1.0728,
      "step": 1430
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.706531516175813,
      "learning_rate": 3.946782899779667e-05,
      "loss": 1.1324,
      "step": 1435
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.8815003834638072,
      "learning_rate": 3.875742443224451e-05,
      "loss": 1.1515,
      "step": 1440
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.6975354892323637,
      "learning_rate": 3.805193155863247e-05,
      "loss": 1.0919,
      "step": 1445
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.7405117357753763,
      "learning_rate": 3.7351406957919636e-05,
      "loss": 1.1764,
      "step": 1450
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.7778411867278201,
      "learning_rate": 3.665590681260658e-05,
      "loss": 1.1504,
      "step": 1455
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.8723639773246538,
      "learning_rate": 3.59654869022294e-05,
      "loss": 1.2015,
      "step": 1460
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.7314301031903592,
      "learning_rate": 3.5280202598886324e-05,
      "loss": 1.1912,
      "step": 1465
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.78382821788559,
      "learning_rate": 3.4600108862796796e-05,
      "loss": 1.1531,
      "step": 1470
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.746538778895017,
      "learning_rate": 3.392526023789349e-05,
      "loss": 1.1861,
      "step": 1475
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.7335920604121904,
      "learning_rate": 3.325571084744803e-05,
      "loss": 1.1032,
      "step": 1480
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.8181126090151531,
      "learning_rate": 3.259151438973024e-05,
      "loss": 1.2354,
      "step": 1485
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.8813117253151651,
      "learning_rate": 3.1932724133701344e-05,
      "loss": 1.0998,
      "step": 1490
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.7703731279590557,
      "learning_rate": 3.1279392914742046e-05,
      "loss": 1.1732,
      "step": 1495
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.8022002883485505,
      "learning_rate": 3.06315731304148e-05,
      "loss": 1.2005,
      "step": 1500
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.7718995592442958,
      "learning_rate": 2.998931673626175e-05,
      "loss": 1.0812,
      "step": 1505
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.7753353561939175,
      "learning_rate": 2.935267524163774e-05,
      "loss": 1.2016,
      "step": 1510
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.8453046609631811,
      "learning_rate": 2.872169970557913e-05,
      "loss": 1.2015,
      "step": 1515
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.788689710824734,
      "learning_rate": 2.8096440732709083e-05,
      "loss": 1.2301,
      "step": 1520
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.9498609267061067,
      "learning_rate": 2.7476948469178887e-05,
      "loss": 1.1721,
      "step": 1525
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.2548603296587735,
      "learning_rate": 2.6863272598646106e-05,
      "loss": 1.1309,
      "step": 1530
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.8088505633219847,
      "learning_rate": 2.625546233829016e-05,
      "loss": 1.1924,
      "step": 1535
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.8053156607642844,
      "learning_rate": 2.5653566434864928e-05,
      "loss": 1.1771,
      "step": 1540
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.0076517594688355,
      "learning_rate": 2.5057633160789184e-05,
      "loss": 1.1547,
      "step": 1545
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.7807995790198267,
      "learning_rate": 2.446771031027527e-05,
      "loss": 1.2168,
      "step": 1550
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.7811855005276076,
      "learning_rate": 2.3883845195495878e-05,
      "loss": 1.2407,
      "step": 1555
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.7607446953536656,
      "learning_rate": 2.330608464278953e-05,
      "loss": 1.1777,
      "step": 1560
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.8030643238657801,
      "learning_rate": 2.273447498890521e-05,
      "loss": 1.2321,
      "step": 1565
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.707300837474353,
      "learning_rate": 2.2169062077286075e-05,
      "loss": 1.103,
      "step": 1570
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.7398162416161335,
      "learning_rate": 2.1609891254392678e-05,
      "loss": 1.082,
      "step": 1575
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.8829809189432961,
      "learning_rate": 2.1057007366066373e-05,
      "loss": 1.1863,
      "step": 1580
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.9036595461566347,
      "learning_rate": 2.0510454753932395e-05,
      "loss": 1.1125,
      "step": 1585
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.6793566054364535,
      "learning_rate": 1.9970277251843862e-05,
      "loss": 1.1453,
      "step": 1590
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.7465371914976613,
      "learning_rate": 1.9436518182366158e-05,
      "loss": 1.1914,
      "step": 1595
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.6678427111685153,
      "learning_rate": 1.8909220353302392e-05,
      "loss": 1.1865,
      "step": 1600
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.8490303048462068,
      "learning_rate": 1.838842605426031e-05,
      "loss": 1.0754,
      "step": 1605
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.6832719345179151,
      "learning_rate": 1.7874177053260598e-05,
      "loss": 1.0879,
      "step": 1610
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.9649690559163364,
      "learning_rate": 1.736651459338695e-05,
      "loss": 1.0977,
      "step": 1615
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.6985183059667567,
      "learning_rate": 1.6865479389478545e-05,
      "loss": 1.1476,
      "step": 1620
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.6282090702067379,
      "learning_rate": 1.6371111624864543e-05,
      "loss": 1.1299,
      "step": 1625
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.8512442207678547,
      "learning_rate": 1.5883450948141377e-05,
      "loss": 1.1309,
      "step": 1630
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.7151617030884643,
      "learning_rate": 1.540253646999299e-05,
      "loss": 1.1936,
      "step": 1635
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.7567039565297413,
      "learning_rate": 1.4928406760054059e-05,
      "loss": 1.1441,
      "step": 1640
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.8464089358888328,
      "learning_rate": 1.4461099843816684e-05,
      "loss": 1.145,
      "step": 1645
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.7170403140830546,
      "learning_rate": 1.4000653199580782e-05,
      "loss": 1.1887,
      "step": 1650
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.7735313453026896,
      "learning_rate": 1.3547103755448287e-05,
      "loss": 1.1786,
      "step": 1655
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.7081409924001252,
      "learning_rate": 1.3100487886361379e-05,
      "loss": 1.2186,
      "step": 1660
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.6796669559142102,
      "learning_rate": 1.266084141118542e-05,
      "loss": 1.1473,
      "step": 1665
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.7301679940106633,
      "learning_rate": 1.2228199589835999e-05,
      "loss": 1.1798,
      "step": 1670
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.838436487459379,
      "learning_rate": 1.1802597120451286e-05,
      "loss": 1.1464,
      "step": 1675
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.8915729376099731,
      "learning_rate": 1.1384068136609105e-05,
      "loss": 1.1965,
      "step": 1680
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.6967288554874705,
      "learning_rate": 1.0972646204589377e-05,
      "loss": 1.1211,
      "step": 1685
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.8038608019365008,
      "learning_rate": 1.0568364320682178e-05,
      "loss": 1.1961,
      "step": 1690
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.8793813680766012,
      "learning_rate": 1.0171254908541372e-05,
      "loss": 1.0912,
      "step": 1695
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.7353307810953187,
      "learning_rate": 9.781349816584162e-06,
      "loss": 1.1382,
      "step": 1700
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.7264376523774766,
      "learning_rate": 9.398680315436903e-06,
      "loss": 1.1503,
      "step": 1705
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.8294378899418743,
      "learning_rate": 9.023277095427173e-06,
      "loss": 1.1001,
      "step": 1710
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.6847190426952564,
      "learning_rate": 8.655170264122303e-06,
      "loss": 1.1837,
      "step": 1715
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.7123063489929077,
      "learning_rate": 8.294389343914899e-06,
      "loss": 1.1665,
      "step": 1720
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.7161490463804174,
      "learning_rate": 7.940963269654922e-06,
      "loss": 1.2128,
      "step": 1725
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.6942289082099549,
      "learning_rate": 7.594920386329252e-06,
      "loss": 1.1516,
      "step": 1730
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.7000185077826192,
      "learning_rate": 7.256288446788362e-06,
      "loss": 1.0752,
      "step": 1735
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.8141301323800174,
      "learning_rate": 6.925094609520455e-06,
      "loss": 1.1663,
      "step": 1740
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.7801337376407085,
      "learning_rate": 6.601365436473439e-06,
      "loss": 1.1264,
      "step": 1745
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.826763084695081,
      "learning_rate": 6.2851268909245865e-06,
      "loss": 1.1076,
      "step": 1750
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.7778658882619575,
      "learning_rate": 5.976404335398256e-06,
      "loss": 1.185,
      "step": 1755
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.5899787648756437,
      "learning_rate": 5.675222529631841e-06,
      "loss": 1.1383,
      "step": 1760
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.7296161663517282,
      "learning_rate": 5.381605628590003e-06,
      "loss": 1.1406,
      "step": 1765
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.7308791705621759,
      "learning_rate": 5.095577180527378e-06,
      "loss": 1.1641,
      "step": 1770
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.7165596394167716,
      "learning_rate": 4.817160125100106e-06,
      "loss": 1.23,
      "step": 1775
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.8126491682868108,
      "learning_rate": 4.546376791525975e-06,
      "loss": 1.1496,
      "step": 1780
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.6916149503964037,
      "learning_rate": 4.2832488967935795e-06,
      "loss": 1.1613,
      "step": 1785
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.6713280206828187,
      "learning_rate": 4.02779754392072e-06,
      "loss": 1.1778,
      "step": 1790
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.8687927248580933,
      "learning_rate": 3.780043220261764e-06,
      "loss": 1.1947,
      "step": 1795
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.9156585367622113,
      "learning_rate": 3.540005795864709e-06,
      "loss": 1.0788,
      "step": 1800
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.8066844554106848,
      "learning_rate": 3.3077045218775192e-06,
      "loss": 1.2058,
      "step": 1805
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.8260344643458865,
      "learning_rate": 3.0831580290041184e-06,
      "loss": 1.1859,
      "step": 1810
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.6728735284202596,
      "learning_rate": 2.8663843260103074e-06,
      "loss": 1.1548,
      "step": 1815
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.8464248337059433,
      "learning_rate": 2.6574007982793857e-06,
      "loss": 1.1424,
      "step": 1820
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.7582737017596137,
      "learning_rate": 2.456224206417812e-06,
      "loss": 1.1132,
      "step": 1825
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.7570508766628901,
      "learning_rate": 2.262870684911045e-06,
      "loss": 1.1916,
      "step": 1830
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.6948287618722808,
      "learning_rate": 2.0773557408295343e-06,
      "loss": 1.1384,
      "step": 1835
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.7078737172323022,
      "learning_rate": 1.8996942525850047e-06,
      "loss": 1.1501,
      "step": 1840
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.767530275514686,
      "learning_rate": 1.7299004687372665e-06,
      "loss": 1.1267,
      "step": 1845
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.7858189878444168,
      "learning_rate": 1.5679880068514174e-06,
      "loss": 1.1056,
      "step": 1850
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.8212356253611762,
      "learning_rate": 1.4139698524057165e-06,
      "loss": 1.1939,
      "step": 1855
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.7961839613413594,
      "learning_rate": 1.2678583577501624e-06,
      "loss": 1.1662,
      "step": 1860
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.850673264738278,
      "learning_rate": 1.1296652411158182e-06,
      "loss": 1.17,
      "step": 1865
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.7063089618156579,
      "learning_rate": 9.994015856749527e-07,
      "loss": 1.1253,
      "step": 1870
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.7573170758624328,
      "learning_rate": 8.770778386522627e-07,
      "loss": 1.0771,
      "step": 1875
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.8837077908686763,
      "learning_rate": 7.627038104869199e-07,
      "loss": 1.1685,
      "step": 1880
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.7524180298657765,
      "learning_rate": 6.562886740457797e-07,
      "loss": 1.1406,
      "step": 1885
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.7266831937495886,
      "learning_rate": 5.578409638877457e-07,
      "loss": 1.0525,
      "step": 1890
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.8486209379295809,
      "learning_rate": 4.6736857557925227e-07,
      "loss": 1.1348,
      "step": 1895
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.7660696151853608,
      "learning_rate": 3.8487876506106966e-07,
      "loss": 1.2252,
      "step": 1900
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.7491200820319253,
      "learning_rate": 3.1037814806634815e-07,
      "loss": 1.1097,
      "step": 1905
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.7181576551220783,
      "learning_rate": 2.43872699590042e-07,
      "loss": 1.0703,
      "step": 1910
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.7330097823119024,
      "learning_rate": 1.8536775340970425e-07,
      "loss": 1.1908,
      "step": 1915
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.8782265197206924,
      "learning_rate": 1.348680016577397e-07,
      "loss": 1.1906,
      "step": 1920
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.7862482683388468,
      "learning_rate": 9.237749444505062e-08,
      "loss": 1.1362,
      "step": 1925
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.7669692568908726,
      "learning_rate": 5.7899639536251883e-08,
      "loss": 1.1689,
      "step": 1930
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.8047883869461449,
      "learning_rate": 3.143720207635648e-08,
      "loss": 1.1729,
      "step": 1935
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.9030824345194336,
      "learning_rate": 1.299230436898613e-08,
      "loss": 1.196,
      "step": 1940
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.7879987601052318,
      "learning_rate": 2.566425706218567e-09,
      "loss": 1.0957,
      "step": 1945
    },
    {
      "epoch": 1.0,
      "eval_loss": NaN,
      "eval_runtime": 1989.8167,
      "eval_samples_per_second": 3.484,
      "eval_steps_per_second": 0.871,
      "step": 1949
    },
    {
      "epoch": 1.0,
      "step": 1949,
      "total_flos": 1.2135540828143616e+16,
      "train_loss": 1.2706995834871218,
      "train_runtime": 18075.5808,
      "train_samples_per_second": 3.45,
      "train_steps_per_second": 0.108
    }
  ],
  "logging_steps": 5,
  "max_steps": 1949,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 1.2135540828143616e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}